Testing monsoonal controls on bedrock river incision in the Himalaya and Eastern Tibet with a stochastic-threshold stream power model

摘要

Be-10-derived catchment average erosion rates from the Himalaya and Eastern Tibet show different relationships with normalized channel steepness index (k(sn)), suggesting differences in erosional efficiency of bedrock river incision. We used a threshold stream power model (SPM) combined with a stochastic distribution of discharges to explore the extent to which this observation can be explained by differences in the mean and variability of discharge between the two regions. Based on the analysis of 199 daily discharge records (record lengths 3-45years; average 18.5years), we parameterized monsoonal discharge with a weighted sum of two inverse gamma distributions. During both high- and low-flow conditions, annual and interannual discharge variabilities are similarly low in each region. Channel widths for 36 rivers indicate, on average, 25% wider streams in Eastern Tibet than in the Himalaya. Because most catchments with Be-10 data are not gauged, we constrained mean annual discharge in these catchments using gridded precipitation data sets that we calibrated to the available discharge records. Comparing Be-10-derived with modeled erosion rates, the stochastic-threshold SPM explains regional differences better than a simple SPM based on drainage area or mean annual runoff. Systematic differences at small k(sn) values can be reconciled with k(sn)-dependent erosion thresholds, whereas substantial scatter for high k(sn) values persists, likely due to methodological limitations. Sensitivity analysis of the stochastic-threshold SPM calibrated to the Himalaya indicates that changes in the duration or strength of summer monsoon precipitation have the largest effect on erosional efficiency, while changes in monsoonal discharge variability have almost no effect. The modeling approach presented in this study can in principle be used to assess the impact of precipitation changes on erosion.